← All problems

Keller problem for the lowest positive eigenvalue of Dirac operators with Lorentz scalar potentials

Let Dm=−iα⋅∇+mβ\mathcal{D}_m=-i\boldsymbol{\alpha}\cdot\nabla+m\beta be the free Dirac operator on Rd\mathbb{R}^d, and consider Dm−Vβ\mathcal{D}_m-V\beta with Lorentz scalar potential VV. For a prescribed LpL^p norm, determine the minimum of the lowest positive eigenvalue λ1+(Dm−Vβ)\lambda_1^+(\mathcal{D}_m-V\beta) over the admissible potentials VV satisfying ∥V∥Lp=c\lVert V\rVert_{L^p}=c. In particular, the unresolved three-dimensional zero-mode question asks whether, for d=3d=3, there exists an admissible potential VV such that 00 is an eigenvalue of Dm−Vβ\mathcal{D}_m-V\beta.

References

Progress summary

Refreshed
Open · progress reported

A new preprint establishes optimizers and rules out zero modes in low dimensions, but leaves the three-dimensional conjecture unresolved.

The problem concerns optimizing the lowest positive eigenvalue of Dirac operators with Lorentz scalar potentials under a prescribed norm. The new work connects the variational problem with solitary-wave equations and examines whether zero modes can occur.

October 2026 development

David Gontier and Hanne Van Den Bosch claim existence of optimal potentials, rule out zero modes in dimensions one and two, and present numerical evidence for nonradial optimizers in dimension two. The conjectured existence of zero modes in dimension three is not proved.

Current status (as of October 2026): existence of optimizers and exclusion of zero modes in dimensions one and two are claimed, while the three-dimensional zero-mode question and the two-dimensional shape conclusions remain unverified.

Sources

Solutions 0

No solutions have been posted yet.